Image deblurring has been studied for decades, but it is still an open problem. Our work focuses on the general single image deblurring, and face image deblurring. Facial expression analysis can be applied to the restored images.

General image deblurring with noise handling: a new method for handling noise in blind image deconvolution is proposed based on new theoretical and practical insights. Our key observation is that applying a directional low-pass filter to the input image greatly reduces the noise level, while preserving the blur information in the orthogonal direction to the filter. Thus, our method applies a series of directional filters at different orientations to the input image, and estimates an accurate Radon transform of the blur kernel from each filtered image. Finally, we reconstruct the blur kernel using inverse Radon transform.

Face image deblurring: the human face is one of the most essential focuses in many applications. So it is important to refine the face images from the blurry images for further analysis. In our method, face landmark detection is first employed to locate the edges of the faces, and then reconstruct an initial gradient map of the face. A L0-regularized gradient prior is utilized in the following iterative deblurring process until convergence.

Facial expression analysis: we present an idea to analyze facial expression by exploring some common and specific information among different expressions. A two-stage multi-task sparse learning (MTSL) framework is proposed to efficiently locate those discriminative patches. With the learned patches, our method can achieve better performance in facial expression recognition.